refactor: Move Streams into VmStateMut

[nyunyunyunyu]

Previously some chips hold Arc<Mutex<Streams>> to read from Streams. Move Streams into VmStateMut so it has all data about VM state.

[codspeed-hq]

CodSpeed Walltime Performance Report

Merging #1707 will improve performances by 13.62%

_{Comparing refactor/streams (257ddb3) with feat/new-execution (82f7ff9)}

⚠️ Unknown Walltime execution environment detected

Using the Walltime instrument on standard Hosted Runners will lead to inconsistent data.

For the most accurate results, we recommend using CodSpeed Macro Runners: bare-metal machines fine-tuned for performance measurement consistency.

Summary

⚡ 8 improvements
✅ 2 untouched benchmarks

Benchmarks breakdown

	Benchmark	BASE	HEAD	Change
⚡	benchmark_execute[factorial_iterative_u256]	558.6 ms	506 ms	+10.4%
⚡	benchmark_execute[fibonacci_iterative]	560.8 ms	503.2 ms	+11.45%
⚡	benchmark_execute[fibonacci_recursive]	538.8 ms	482.1 ms	+11.77%
⚡	benchmark_execute[keccak256]	566.4 ms	498.5 ms	+13.62%
⚡	benchmark_execute[keccak256_iter]	560.9 ms	500.4 ms	+12.08%
⚡	benchmark_execute[quicksort]	562.1 ms	508.9 ms	+10.47%
⚡	benchmark_execute[sha256]	559.2 ms	500.5 ms	+11.74%
⚡	benchmark_execute[sha256_iter]	538.6 ms	489.1 ms	+10.11%

[codspeed-hq]

CodSpeed Instrumentation Performance Report

Merging #1707 will not alter performance

_{Comparing refactor/streams (257ddb3) with feat/new-execution (82f7ff9)}

Summary

✅ 10 untouched benchmarks

[github-actions]

group	app.proof_time_ms	app.cycles	app.cells_used	leaf.proof_time_ms	leaf.cycles	leaf.cells_used
verify_fibair	(-55 [-4.7%]) 1,124	(+11347 [+3.5%]) 334,023	(+336128 [+1.9%]) 17,675,984	-	-	-
fibonacci	(-459 [-17.7%]) 2,127	1,500,277	50,589,503	-	-	-
regex	(-882 [-11.5%]) 6,797	4,165,432	166,519,456	-	-	-
ecrecover	(+40 [+2.9%]) 1,436	289,547	(+43030 [+0.3%]) 14,513,216	-	-	-
pairing	(-140 [-3.1%]) 4,359	1,820,436	(+346586 [+0.4%]) 96,178,993	-	-	-

Commit: 257ddb3

Benchmark Workflow

[jonathanpwang]

To check, the logic to move into VmStateMut directly instead of Ctx is because it is shared regardless of Ctx, so might as well make it global?

[nyunyunyunyu]

To check, the logic to move into VmStateMut directly instead of Ctx is because it is shared regardless of Ctx, so might as well make it global?

Yes, it's not specific to any execution mode

[jonathanpwang]

LGTM except I don't understand why the metrics thing changed